Initial concept

• Expand current features to include gffcompare-output-gtf-file and reference-gtf-file. The arguments could be -i or --input for gffcompare-output-gtf and -r or --reference
• Content from each file is required
• Content from each file is stored into a separate database with same conditionalities as currently is being done

gffutils provides tools for storing gtf-file content into a relation database (sqlite3). This tool can be used in the creation of the wanted feature.

See: gffutils

Related user story: #1

Related user story: #1

Details

The current sample-data consists only of data related to one transcript. It's good to keep the test dataset small, but it should also be more extensive. Create a new dataset with 3-4 transcripts. Some unittests will break. Fix them.

Update

No more need for more extensive sample test. Closing issue.

Task

• Add argument -r or --reference
• Argument is required

Related user story: #14

Related user story: #1

Tasks

• Add logging library
• Refactor logging into categories (info, debug, warning,...)
• Add argument for enabling debug logging
• Add argument for writing log information to a file (append, no overwrite)

Initial concept

A user can use an argument to specify an output file. The argument is -o or --output. If no argument is given, output will be given as stdout.

Revision

After consideration the optimal solution seemed to be to store stdout to a file as a default. This has been implemented. Closing issue.

Task

• Expand current model to manage creation / uploading of two databases

Related user story: #14

Details

• Initial tests added. An open issue is that the tests are more like integration tests testing the whole application. This might cover underlying issues.
• Consider mocking called classes and methods to more precisely test that everything is called properly.

Update 3rd of July 2023

• Code refactored, new unittests written
• Persisting probel is that testing compana.py covers parts of code that might not have unittests.
• Reserve time for writing missing unittests after this task is finished. Create new issues as needed.

Related user story: #15

Related user story: #50

Functionality

• User provides full path to gtf file as an CLI argument.
• A database is created by using gffutils
• location of database is the path of the gtf file. Database filename is <gtf-filename w/o extension>-ca.db

Related user story: #1

Details

• Write documentation explaining actions in alignment parsing and indel identification
• Write documentation for BAM-processing
• Write documentation for the pipeline

Task

• Add a new argument c or class-code
• Argument is optional (at this stage)

Related user story: #15

Details

• Add strand to offset output information
• First find out what strand is)

Related user story: #15

Found a non-common instance where offsets are not mapped correctly in situations where there are mismatching alignments in both reference- and aligned-data.

Intuitively the offset computation works, but a formal definition and explanation should be formulated. Add the definition and explanation into the documentation. Consider additionally creating a proof.

Related user story: #15

Initial concept

Initial plan is to calculate the offset for exons. The steps for this could be:

• user can give desired class code as an argument.
• gffcompare-output-GTF file: access a transcript containing a selected access code and extract the gene_id (to locate the exons in the reference file related to the gene)
• gffcompare-output-GTF-file: access exons related to that transcript
• compare the location of matching exons in both files (reference-GTF-file and gffcompare-output-GTF-file) and calculate the offset for each exon as a tuple (s, e), in which s = offset at start location and e = offset at end location

The output should have a row for each transcript with the following items:

• Transcript identification
• A list of offsets

An example:

<transcript id>;[ (s_1, e_1), (s_2, e_2), ..., (s_n, e_n)]

The output format and delimiter(s) need to be planned more

Key functionalities

• by giving an argument -s or --stats user is shown statistics on the given gtf-file. Syntax

compana.py -i <filename> -s

Output

The output should be

Class code           n
<code>              <n>
<code>              <n>

Details

Lot's of new code was written in sprint 5, but the quality is still weak. Improve code quality and write unittests for new code.

Initial idea

• The transcript comparison should be it's own function
• The function takes two lists as an input
• The function returns a list with smallest offsets computed for each value on the list.

Definitions

• offset: an exon can have an offset in the start index and/or in the end index. The offset is a two element tuple with the following values (start_idx - reference.start_idx, end_idx - reference.end_idx)
• total offset: the absolute distance of offset in start and end: total_offset = |start_offset| + | end_offset |
• smallest offset: is the smallest total offset

Related user story: #15

Initial concept

A user should be able to store data from CLI. The workflow could be:

• User calls the application with a command line instruction (e.g. python3 <filename> <arguments>
• As an argument user provides file directory
• output is a db-file in the directory where the given file is located.

Directory example

The given input-file does not need to be located in the same directory as the utility.

|
└─── util 
            util.py
|
|
└─── file location
            <file>.gtf
            <file>-ca.db

The initial idea for the fuctionality:

flowchart LR
    A{Start util} --> B[db exists?]
    B -->|Yes| C{Remove db?}
    C -->|No| D[End]
    C -->|Yes| E[delete db]
    E --> F[Run util]
    B -->|No| F
    F --> D

pyfaidx included in requirements.txt.

Related user story: #50

Details

Some initial code exists in the file cigar-parser.py. However, pysam's get_aligned_pairs might prove more effective for this use case. First get familiar with pysam's functionalities and then implement a solution for the given task. See documentation for sprint-6 for more details.

Update 20.6.2022: For now we will leave mismatches out and focus on indels.

Finish first: #84

Details

Create an initial concept for a data structure and include it in the general information -documentation.

Related user story: #50

• Install and import gffutils
• Add gffutils to requirements.txt

Related user story: #1